Estimating impact of puddling, tillage and residue management on wheat seedling emergence and growth...

www.elsevier.com/locate/still

Soil & Tillage Research 87 (2006) 119–130

Estimating impact of puddling, tillage and residue management

on wheat (Triticum aestivum, L.) seedling emergence and growth

in a rice–wheat system using nonlinear regression models

M. Mohanty *, D.K. Painuli, A.K. Misra, K.K. Bandyopadhyaya, P.K. Ghosh

Indian Institute of Soil Science, Indian Council of Agricultural Research, Nabi Bagh,

Berasia Road, Bhopal 462038, Madhya Pradesh, India

Received 3 November 2004; received in revised form 2 March 2005; accepted 14 March 2005

Abstract

Puddling is known to increase the yield of rice due to the creation of suitable physical environment that favours growth of the

crop. However, in rice–wheat system, wheat yield has been reported to decrease due to the deterioration of soil structure caused

by puddling in rice. This affects seedling emergence in wheat. Seedling emergence model that predicts seedling emergence and

early growth of wheat can be used to estimate the major effects of different tillage and residue management practices on seedling

shoot and root growth. Output from such a model can be used to initialize crop growth models under diverse soil and climatic

conditions in which the variations in crop establishment are often poorly taken into account The information on predicting wheat

(Triticum aestivum, L.) seedling emergence and growth using nonlinear regression models under rice–wheat cropping system is

essential as early emergence and growth of seedlings affect the grain yield. A study was undertaken to assess the residual effect

of puddling in rice (no puddling i.e. direct seeding, puddling by four passes of 5 hp power tiller and puddling by eight passes of

5 hp power tiller) and direct effect of different tillage (conventional and zero tillage) and residue (residue-retained and removed)

management practices on wheat seedling emergence and growth in rice–wheat system on a Vertisol of Central India. Wheat

seedling emergence was maximum where rice was direct seeded, and wheat was grown under conventional tillage with residue

retained at the surface. Prediction of wheat seedling emergence by the [France, J., Thornley, J.H.M., 1984. Mathematical Models

in Agriculture and Related Sciences. Butterworth, London] model, shoot growth by the Logistic and Gompertz models, and root

growth by Monomolecular model was attempted. Time to reach 50% emergence as predicted by the [France, J., Thornley,

J.H.M., 1984. Mathematical Models in Agriculture and Related Sciences. Butterworth, London] model was closer to the

observed emergence data. The nonlinear regression models study indicated that the Logistic model predicted the shoot growth of

wheat under different tillage and residue management practices better than the Gompertz model. Whereas, for root growth the

Monomolecular model fitted well with the experimental data.

# 2005 Elsevier B.V. All rights reserved.

Keywords: Puddling; Zero tillage; Wheat seedling; Seedling growth and emergence; Model; Vertisol; Rice–wheat cropping system

* Corresponding author. Tel.: +91 755 2730970; fax: +91 755 2733310.

E-mail address: [email protected] (M. Mohanty).

0167-1987/$ – see front matter # 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.still.2005.03.002

M. Mohanty et al. / Soil & Tillage Research 87 (2006) 119–130120

1. Introduction

Tillage is practiced in soils for controlling weeds,

breaking crusts (improving water entry), increasing

surface roughness (assisting water storage) and

preparing a seedbed. The type of tillage method to

be practiced, however, depends upon the soil type and

the climate of the area. For Vertisols and vertic soils

preparation of seedbed, improvement of soil structure

and conservation of soil and water are of great concern

(Coughlan et al., 1989). It is suggested that to enhance

the productivity of these soils best tillage systems need

to be practiced. Thomas et al. (1990) reported that for

improving soil water storage (0–1.8 m) in a Vertisol,

use of crop residue mulch was effective. Various

techniques viz., use of crop residue mulch, deep

tillage, raised bed in the form of ridges or broad beds

and furrow systems have been recommended for these

soils (Anonymous, 1995). However, zero tillage (ZT)

has been widely acclaimed as highly effective practice

for conservation of soil and water in Vertisols as,

compared with conventional tillage (CT), it enhances

water conservation through improved infiltration and

reduced evaporation (Unger and Ordief, 1989). Tillage

affects mechanical characteristics of the seedbed

considerably and so can influence crop emergence.

According to Leslie (1965), crop emergence is

commonly a major problem in Vertisols due to

inadequate soil moisture and poor structure of the

seedbed. Maintenance of residue on soil surface

through ZT has been shown to benefit some Vertisols

due to improved soil physical properties in north-

eastern Australia (Freebairn et al., 1986a, 1986b;

Sallaway et al., 1990). While Radford and Nielsen

(1983) have reported that stubble retention may

improve seedling emergence in Vertisols whereas

Thomas et al. (1990) have reported that zero and

reduce tillage with stubble retention decreased

seedling emergence at the same site. Chastain et al.

(1994) reported that high levels of residue cover

reduced the emergence rate of wheat probably by

reducing seed soil contact. Swan et al. (1996) reported

that removing excessive plant residue from the seed

row increased the germination and emergence rate of

maize. There is a great need to predict the effect of soil

tillage, seedbed preparation, and sowing techniques on

crop establishment. These technical changes are costly

for growers, but their effects are not easy to predict.

They cause changes in physical condition of the

seedbed and in seed placement (Aubertot et al., 1999),

which interact with seed treatment, seed character-

istics, and cultivars to determine the features of the

crop stand.

Initial plant population has direct bearing on crop

yield and seedling emergence influences the initial

plant population. Gan et al. (1992) reported that the

wheat plants that emerged early contribute more

towards crop yield than those that emerged later.

Thus, desirable crop yields are achieved by providing

seeds with an environment that encourages early

germination and emergence. Several authors have

emphasized the importance of analyzing the stand

establishment process and have shown that the main

factors affecting germination, seedling emergence

and plant establishment are associated with the

mechanical characteristics of the seedbed (Jensen,

1971; Blacklow, 1972; Schneider and Gupta, 1985).

Tillage influences bulk density, penetration resis-

tance, aggregate mean weight diameter and surface

roughness (Carman, 1996).

Normally the crop stand of wheat in rice–wheat

cropping system is lower after rice because of the

destruction of soil structure during puddling in rice

(Sharma et al., 1995). Puddling can be defined as the

process of breaking down soil aggregates into uniform

mud, accomplished by applying mechanical force to the

soil at high moisture content for transplanting of rice

seedlings. Conventional experiments comparing the

plant populations and final yields of different manage-

ment techniques are time consuming and expensive. A

seedling emergence model that predicts seedling

emergence and early growth of wheat could be used

to estimate the major effects of different tillage and

residue management practices on seedling shoot and

root growth. Output from such a model could be used to

initialize crop growth models under diverse soil and

climatic conditions in which the variations in crop

establishment are often poorly taken into account

(Brisson et al., 1998; Guerif et al., 1998). Mohanty and

Painuli (2004) have reported the effect of tillage and

residue on emergence and growth rice seedlings using

nonlinear regression models in a Vertisol. However,

such information on wheat seedling emergence and

growth is very limited. Though some models have been

developed for wheat and other crops for emergence

based on nonlinear regression models (Bouaziz and

M. Mohanty et al. / Soil & Tillage Research 87 (2006) 119–130 121

Fig. 1. Rainfall distribution during the 3 years of experimentation.

Bruckler, 1989a, 1989b, 1989c; Finch Savage and

Phelps, 1993; Mullins et al., 1996), the effects of tillage

and residue on seedling growth are not taken into

account in these models. Forcella et al. (2000), has

reported a comparison of empirical and mechanistic

model on seedling emergence. They reported that

though mechanistic models takes into account soil, seed

and climatic environment into consideration, but they

are the most difficult models to develop. Creation and

use of simpler empirical models, which also employ

microclimate and soil factors for predictions, may

provide satisfactory predictions of seedling emergence

until better mechanistic models are developed. A

model predicting seedling emergence in sugarbeet

has been developed taking into account the soil tillage

and sowing operations by Durr et al. (2001). They

predicted the emergence time in sugarbeet by using

Weibull function and early growth by exponential

function (Durr and Boiffin, 1995). Gompertz, Weilbull,

Richards, and Logistic functions have been used

successfully in predicting the cumulative relative seed-

ling emergence (CRE) of some weed species (Brown

and Mayer, 1988). Some researchers have used compli-

cated extensions of these functions, such as double

Gompertz curve (Kremer and Lotz, 1998). Most of the

emergence model studies are confined to weeds only.

Use of thermal time in CRE prediction of weeds

includes Digitaria ischmaeum (Fidanza et al., 1996), S.

halepense (Benech Arnold et al., 1990) and 15 other

species in computer software called WeedCast (For-

cella, 1998). Some researchers have used calendar days

as time variable in CRE models (Cussans et al., 1996;

Vleeshouwers, 1997). An empirical model, represent-

ing emergence over time (calendar days) under field

observation of Digitaria Sanguinalis was reported by

King and Oliver (1994). The effective empirical models

that combine soil temperature and soil water potential

(Finch Savage and Phelps, 1993) are supportive to

increasing the information-richness of agronomic

management decisions, which have been proven to

enhance the timeliness and cost effectiveness of

standard management operations (Forcella, 1998).

The present study was undertaken to determine the

effect of puddling, stillage and residue management on

emergence and growth of wheat seedling shoot and root

in a rice–wheat system on a Vertisol as predicted by

suitable nonlinear regression models. In this study some

pertinent nonlinear regression models were evaluated to

predict the wheat seedling emergence and growth in

field condition. The shoot height/length and rooting

depth denote the growth of wheat seedlings in the

present study.

2. Materials and methods

2.1. Site and soil

The field experiment was conducted during the

winter seasons of the years 2000–2002 at the experi-

mental farm of the Indian Institute of Soil Science,

Bhopal, Madhya Pradesh, India, located at 238180Nlatitude, 778240E longitude, 485 m above mean sea

level. The soil is an Entic Chromusterts with 52% clay,

30% silt, 18% sand, pH 7.8, cation exchange capacity

49 cmol (p+) kg�1 and organic carbon 4.6 g kg�1 in the

0–15 cm layer. Bhopal receives on an average 1200 mm

annual rainfall. Most of it is received during the month

of June–September (Fig. 1). However, the distribution

of rainfall is very erratic in this region.

2.2. Tillage and residue levels

The treatment details for rice (Oryza sativa, L.) and

wheat (Triticum aestivum, L.) are presented as follows.

In the first year, Rice (cv IR 36) was grown during the


Table 1

Empirical constants a, b, coefficient of determinations (R2), T0.5 and MER for emergence of wheat seedlings as estimated from Logistic model

(France and Thornley, 1984) for various treatments

Treatments a b R2 T0.5 MER

Predicted Observed

P0 3.87 0.92 0.99 4.20 3.75 10.60

P1 3.85 0.90 0.98 4.27 4.00 5.23

P2 3.55 0.72 0.99 4.93 4.50 5.04

LSD (P = 0.05) 0.05 0.08 0.11 0.25 2.20

Rr 4.13 0.90 0.98 4.58 4.50 4.43

R0 3.60 0.84 0.98 4.28 4.00 5.69

LSD (P = 0.05) 0.10 0.04 NS 0.15 0.55

CT 3.81 0.90 0.99 4.23 4.50 7.39

ZT 3.77 0.84 0.98 4.48 4.00 4.94

LSD (P = 0.05) NS NS NS 0.15 0.39

a: constant of integration; b: emergence rate constant; T0.5: time for 50% emergence; MER: maximum emergence rate.

wet season under three puddling intensities viz., no

puddling i.e. direct seeding (P0), puddling by four

passes of 5 hp power tiller (P1) and puddling by eight

passes of 5 hp power tiller (P2). Power tiller is a semi-

automatic manually operated device with cage wheels

attached, which helps in puddling operation. The

depth of puddled layers was 10.9 cm for P1 and

14.5 cm for P2, respectively. Rice seedlings (25 days

old) were transplanted in P1 and P2 at a row spacing of

20 cm. Fertilizers applied were urea (50% at sowing

and 50% at flowering stage) to provide 90 kg N ha�1,

super phosphate (100% at sowing) to provide

30 kg P ha�1 and muriate of potash (100% at sowing

time) to provide 40 kg K ha�1. At harvesting of rice,

residue of 30 cm height was retained in residue-

retained plots (Rr) simulating mechanical harvesting

and in other plots it was removed (R0) before the

following spring wheat crop. The wheat crop (cv. C

306) was grown under conventional tillage (CT; one

pass of a disc harrow followed by two passes of a duck

foot cultivator) as well as under zero tillage (ZT;

sowing by Pantnagar zero till seed drill in the untilled

soil) with the same levels of residue management

(residue-retained and residue removed). For conven-

tional tillage, conventional seed drill was used

whereas for zero tillage, Pantnagar zero till seed drill

was used. The sowing of wheat was carried out prior to

6 cm of presowing irrigation in dry soil. At wheat

harvest also, residue of 30 cm height (approximately

6000 kg ha�1) was retained in residue-retained plots

(Rr) and removed from other plots (R0) for the

succeeding rice crop. The wheat was sown with a seed

drill at the seed rate of 100 kg ha�1 both for

conventional and zero tillage.

2.3. Plant observations

The observations on wheat seedlings were taken

during third year of experimentation i.e. year 2002.

Seedling emergence was determined by counting the

number of newly emerged seedlings in 1 m length of

row daily with three replications. Relative emergence

(RE) was calculated directly from the final emergence

count.

The RE data was fitted using nonlinear regression

procedures to the Logistic growth model (France and

Thornley, 1984) of the following form

RE ¼ 1

½1 þ expðaþ btÞ� (1)

where a is the constant of integration, b the emergence

rate constant and t the time since sowing. The Logistic

growth model has been used successfully to assess the

effect of environmental conditions on seed germina-

tion (Schimpf et al., 1977). The Logistic growth model

has a point of inflection, at which the rate of emer-

gence reaches a maximum and this occurs at a time

when RE = 0.5M, where M is a parameter describing

the maximum number of seedlings that eventually

emerged (France and Thornley, 1984). The time at

which the point of inflection occurs, also called the

median emergence time (T0.5), is given by:

T0:5 ¼ a

b(2)


The maximum emergence rate (MER) is given by:

MER ¼ Mb

4(3)

For shoot and root growth, two rows were selected

for each treatment. The soil was dug up and the

seedlings were removed daily from day 1 after

sowing till the final observation. After thorough

washing with running tap water for removal of soil,

the length of the shoots and roots were measured

using a scale. For this plants from the 50 cm length

of row were taken out from each treatment with

three replications. Observations were recorded

after 3 years of experimentation in the studied

Vertisol.

2.4. Modeling growth of wheat seedlings

In the absence of a general theoretical equation to

describe seedling growth, three empirical curvilinear

growth functions historically used for analyzing dry

matter accumulation versus time (namely the

Logistic, Gompertz and Monomolecular equations)

were examined for applicability. Since Monomole-

cular model did not fit to the shoot growth of the

experimental curve, this was not discussed for shoot

growth in this study. Similarly, Logistic and

Gompertz models did not predict the rooting depth

well, hence, was not mentioned in Section 3. In this

study only Logistic and Gompertz models were used

for shoot growth and Monomolecular model for root

growth. A common property of these models is that

the length (L) approaches a constant (Lf) as time

approaches infinity. In fact the seedling temporarily

stops increasing in height and increases its number of

leaves just after emergence (2- to 3-leaf stage)

because internodes elongation does not occur until

later and only leaves are appearing as the plant stays

in a rosette form.

2.5. Parameter estimation

On the basis of experimental data related to shoot

and root growth of wheat in different tillage and residue

management practices at a constant temperature, a

nonlinear regression fitting procedure was used to

estimate the parameters of the three functions given

below.

2.5.1. The Logistic model

This is a symmetrical logistic function with an

inflection point:

LðtÞ ¼ L0Lf

L0 þ ½Lf � L0�expð�G 0tÞ(4)

where, L(t) is the length in millimeter of shoot at any

time (t), L0 the length at the onset of growth (t = 0),

and G0 the relative growth rate at time 0 (d�1).

Assuming that L0 = 1 mm just after germination,

the above equation can be written as

LðtÞ ¼ Lf

1 þ ½Lf � 1�expð�G 0tÞ(5)

Thus, only two parameters Lf and G0 were to be

estimated.

2.5.2. The Gompertz model

This is an asymmetrical function with an inflection

point; in this case RGR (relative growth rate) defined

by G, decreases exponentially with time.

LðtÞ ¼ Lf exp½f�lnðLfÞgfexpð�KgtÞg� (6)

where, Kg is the relative growth rate. Only two para-

meters Lf and Kg were to be estimated.

2.5.3. The Monomolecular model

In this mathematical function there is no inflection

point:

LðtÞ ¼ Lf ½1 � expð�KmtÞ� (7)

where Km (d�1) is the proportionality coefficient and

the equation implies that L0 = 0 when t = 0. Two

parameters Lf and Km were to be estimated.

2.6. Model evaluation

Seedling emergence rate and growth of shoot and

root were the variables on which the model predictions

were compared with the observed values. The Logistic

model by France and Thornley (1984) was used for

emergence. Where as both the Logistic and Gompertz

model were used for the growth of shoots of wheat

seedlings and the Monomolecular model was used for

growth of roots. The statistical criteria used to

compare the predicted (Pi) and observed (Oi) values

for growth models have been depicted in Eqs. (8)–(10)

as suggested by Smith et al. (1996).


2.6.1. Modeling efficiency (EF)

It shows the closeness between the observed and

predicted values. The narrower the margin between

them, the better the prediction.

EF ¼Pi¼n

i¼1ðPi � OiÞ2

Pi¼ni¼1ðOi � OÞ2

(8)

Root mean square error (RMSE):

RMSE ¼ 100

O

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiXi¼n

i¼1

ðPi � OiÞn

vuut (9)

2.6.2. The coefficient of residual mass (CRM)

It represents the modeling biasness. CRM describes

whether the model over predict or under predicts the

observed values.

CRM ¼Pi¼n

i¼1 Oi �Pi¼n

i¼1 Pi

� �Pi¼n

i¼1 Oi

(10)

In the above equations, n is the number of times

heights of the seedlings were observed and O is the

mean observed values.

Fig. 2. Observed vs. predicted values for the relative emergence of

wheat in puddling treatments as predicted by Logistic growth model

(year 2002): (A) direct seeded rice; (B) puddling with four passes

power tiller; (C) puddling with eight passes power tiller.

3. Results and discussion

3.1. Seedling emergence parameters

Seedling emergence parameters as influenced by

tillage and residue management practices were

estimated using the Logistic model (France and

Thornley, 1984). The Logistic model fitted well with

the emergence data as evident from the R2 value

(Figs. 2 and 3). Fitting of the relative emergence data

of each treatment to the Logistic model showed that

the parameters a and b were different for all the

treatments (Table 1). There was significant difference

in the values of a and b under different puddling

intensity. P2 registered lower a (3.55) and b (0.72) than

P1 (3.85 and 0.90) and P0 (3.87 and 0.92),

respectively. This showed that puddling in rice had

a significant residual effect depressing the emergence

of the subsequent wheat crop. The values of a and b for

R0 (4.13, 0.90) were significantly higher than Rr (3.60,

0.84). There was no significant difference between CT

and ZT with respect to a and b. The time to reach 50%

emergence as predicted by the model was higher in P2

(4.93 day), R0 (4.58 day) and ZT (4.48 day) than in P1,

Rr and CT, respectively. The higher T0.5 value in P2

was due to delayed emergence (Fig. 4). The number of

seedlings emerged was higher in P0 followed by P1

and P2. This showed puddling in rice significantly


Fig. 3. Observed vs. predicted values for the relative emergence of wheat in different tillage and residue treatments as predicted by Logistic

model (year 2002): (A) conventional tillage; (B) zero tillage; (C) rice residue removed; (D) rice residue retained.

Table 2

Values of statistical criteria for growth models on all tillage and

residue treatments

Treatment Models EF RMSE CRM

P0 Logistic 0.988 7.888 0.018

Gompertz 0.986 7.103 �0.016

P1 Logistic 0.988 6.681 0.015

Gompertz 0.981 7.646 �0.017

P2 Logistic 0.988 6.071 0.014

Gompertz 0.985 8.231 �0.019

CT Logistic 0.989 5.574 0.013

Gompertz 0.985 6.717 �0.015

ZT Logistic 0.993 1.451 0.003

Gompertz 0.985 8.555 �0.020

R0 Logistic 0.994 6.422 0.015

Gompertz 0.986 7.678 �0.018

Rr Logistic 0.995 3.308 0.007

Gompertz 0.980 10.945 �0.025

EF: efficiency of the model; RMSE: root mean square error; CRM:

coefficient of residual mass

decreased the emergence of seedlings in the subse-

quent wheat crop. Similarly, CT and Rr registered

higher seedling numbers than under ZT and R0. The

time to reach T0.5 value was early in case of Rr due to

better soil aeration and moisture content provided by

the residue environment. The earlier attainment of T0.5

in ZT was due to lower ultimate emergence (describ-

ing the maximum number of seedlings that eventually

emerged out). This is in agreement with the results

reported respectively by Radford and Nielsen (1983)

and Thomas et al. (1990). Among the puddling

treatments, the maximum emergence rate (MER) was

obtained under P0 (10.60) as compared to P1 (5.23)

and P2 (5.04) (Table 1). There was no significant

difference in MER between P1 and P2. The reason is

that ploughing of puddle soils after rice results in the

formation of large clods, having high breaking

strength (Sharma et al., 1995) and thus, reduces the

seed–soil contact. The residue retention treatment (Rr)

registered significantly higher MER (5.69) than

residue removed (R0) (4.43). Similarly CT showed

significantly higher MER than ZT. Thus, the results

revealed that tillage and residue management after rice

significantly affected the emergence of wheat seed-

lings. The lower T0.5 value predicted in case of CT

might be due to early emergence, which may be

attributed to better seedbed condition leading to better

seed–soil contact (Thomas et al., 1990; Chastain et al.,

1994).


Fig. 4. Seedlings counts in wheat as influenced by residual effect of puddling (A) and direct effect of tillage (B) and residue of rice (C) (year

2002). Vertical bar represents LSD (P = 0.05) and ns = nonsignificant.

3.2. Comparison of growth models for seedling shoot

and root growth

Three empirical growth functions were compared

to arrive at an appropriate model for predicting shoot

and root growth. For shoot growth curve fitting

generally gave efficiency >0.90 in both the Logistic

and the Gompertz model (Table 2). Comparing the two

models, the Logistic model was considered to be the

best of the two because of the close agreement

between the predicted (EF > 0.98 in most of the cases

and RMSE < 7.89) and the observed values from the

experiment. The Gompertz model was considered to

be inferior because of the higher RMSE values (>6.72

in all the cases). The growth of wheat seedlings as

predicted by predicted by the Gompertz model was

slightly greater than the observed values (CRM having

negative values in all cases) although the shape of the


Fig. 5. Shoot height in wheat seedlings as influenced by residual

effect of puddling (A) and direct effect of tillage (B) and residue of

rice (C) (year 2002). Vertical bar represents LSD (P = 0.05) and

ns = nonsignificant.

Fig. 6. Observed vs. predicted values for the root length of wheat in

puddling treatments as predicted by Monomolecular model (year

2002): (A) direct seeded rice; (B) puddling with four passes power

tiller; (C) puddling with eight passes power tiller.

curve matched the shape of the experimental data

(Fig. 5). Because of this slight over prediction by the

Gompertz model, the Logistic model was considered

best for shoot growth of wheat seedlings. The

Monomolecular model was not taken into account

in the present study, as the shape of the curve did not

match the growth of the shoot. Bouaziz and Bruckler

(1989a) also reported the unsuitability of this model

for predicting the growth of wheat seedlings in the

laboratory. However, for the root growth of wheat

seedlings this model was used instead of the Logistic

and Gompertz model as it fitted well with the

experimental observations (Figs. 6 and 7).

3.3. Estimated parameters of the growth models as

influenced by tillage and residue

Puddling, tillage and residue had different effects

on the Logistic parameters Lf and G0 and the

Monomolecular parameters Lf and Km (Table 3) for


Fig. 7. Observed vs. predicted values for the root length of wheat under tillage and residue treatments as predicted by Monomolecular model

(year 2002): (A) conventional tillage; (B) zero tillage; (C) rice residue removed; (D) rice residue retained.

both shoot and root growth. The Gompertz parameters

Lf and Kg are not reported here as the Logistic

parameters fitted better than the Gompertz model for

shoot growth. The residual effect of puddling had a

significant effect on Lf of shoots in the Logistic model.

The Lf decreased significantly under the Logistic

model with puddling. It also decreased but non-

significantly under ZT and R0 treatments compared

with the CT and Rr treatments. The G0 was

significantly affected by puddling and residue

Table 3

Empirical constants: final length (Lf), G0 and proportional (Km) of the tillag

models

Treatment Shoot

Logistic model

Lf G0 R

P0 284 0.660 0

P1 281 0.588 0

P2 274 0.572 0

LSD (P = 0.05) 2.55 0.010

CT 285 0.578 0

ZT 282 0.566 0

LSD (P = 0.05) NS NS

R0 281 0.536 0

Rr 283 0.583 0

LSD (P = 0.05) NS 0.013

management practices as predicted by Logistic model.

However, the direct effect of tillage on G0 as predicted

by the model was non-significant. Puddling and

residue management practices had significant effect

on Km under the Monomolecular model for root

whereas the effect of tillage was non-significant

(Table 3). Km was affected more by puddling and

residue than tillage as predicted by the models. The

decrease in wheat root growth due to puddling in rice

was due to poor soil physical environment (Oussible

e and residue treatments for shoot and root of wheat under different

Root

Monomolecular model

2 Lf Km R2

.99 47.25 0.090 0.94

.98 48.23 0.082 0.95

.98 47.80 0.075 0.95

NS 0.006

.98 46.12 0.096 0.94

.99 51.58 0.070 0.96

3.25 0.009

.99 57.39 0.083 0.96

.98 51.39 0.083 0.95

2.35 NS


et al., 1992). The parameters G0 and Km were also

significantly affected by puddling, tillage and residue

management. These parameters showed nothing but

the relative growth rates for the models. Thus, it may

be inferred that the residual effect of puddling in rice

and direct effect of tillage and residue management

has significant effect on the shoot and root growth of

wheat seedling.

4. Conclusion

The study showed that puddling in rice had

significant depressing effect on emergence, shoot

and root growth of subsequent wheat seedlings.

Conventional tillage and rice residue retention as

tillage option in wheat favoured seedling emergence

and growth. The nonlinear regression models study

indicated that the Logistic model predicted the

emergence and shoot growth of wheat under different

tillage and residue management practices better than

the Gompertz model. Whereas, for root growth the

Monomolecular model fitted well with the experi-

mental data. These models can further be used as tools

for simulating some important effects of soil tillage

and sowing operations on seedling emergence, which

can be, used as submodel in crop growth simulation

models.

Acknowledgement

The fund provided by National Agricultural

Technology Project PSR No. 31 for this study is duly

acknowledged.

References

Aubertot, J.N., Durr, C., Kieu, K., Richard, G., 1999. Characteriza-

tion of sugarbeet (Beta Vulgaris L.) seedbed structure. Soil Sci.

Soc. Am. J. 63, 1377–1384.

Anonymous, 1995. Tillage Systems in Tropics: Management

Options and Sustainability Implications, in: R. Lal (Ed.), FAO

Soils Bulletin No. 71, pp. 107–120.

Benech Arnold, Ghersa, R., Sanchez, L., Insausti, C.M., 1990. A

mathematical model to predict Sorghum halepense (L.) Pers.

seedling emergence in relation to soil temperature. Weed Res.

30, 91–99.

Blacklow, W.M., 1972. Influence of temperature on germination and

elongation of the radicle and shoot of corn (Zea mays L.). Crop

Sci. 12, 647–650.

Bouaziz, A., Bruckler, L., 1989a. Modelling of wheat imbibition and

germination as influenced by soil physical properties. Soil Sci.

Soc. Am. J. 53, 219–227.

Bouaziz, A., Bruckler, L., 1989b. Modeling wheat seedling growth

and emergence: Seedling growth affected by soil water potential.

Soil Sci. Soc. Am. J. 53, 1832–1838.

Bouaziz, A., Bruckler, L., 1989c. Modelling wheat seedling growth

and emergence: Comparison with field experiments. Soil Sci.

Soc. Am. J. 53, 1838–1846.

Brisson, N., Mary, B., Ripoche, D., Jeufrror, M.H., Ruget, F.,

Nicoullaud, B., Gate, P., Devienne-Barret, F., Antonioletti, R.,

Durr, C., Richard, G., Beaudoin, N., Recous, S., Tayot, X.,

Plenet, D., Cellier, P., Machet, J.M., Maynard, J.M., Delecolle,

R., 1998. A generic model for simulation of crops and their water

and nitrogen balance. I. Theory and parameterization applied to

wheat and corn. Agronomie 18, 311–346.

Brown, R.F., Mayer, D.G., 1988. Representing cumulative germina-

tion. 2. The use of Weilbull function and other empirically

derived curves. Ann. Bot. 61, 127–138.

Carman, K., 1996. Effect of different tillage systems on soil proper-

ties and wheat yield in middle Anatolia. Soil Till. Res. 40 (3/4),

204–207.

Chastain, T.G., Ward, K.J., Wysocki, D.J., 1994. Stand establish-

ment responses of soft white winter wheat to seedbed residue

and seed size. Crop Sci. 35, 213–218.

Coughlan, K.J., Smith, G.D., Yule, D.F., 1989. Soil physical

research for improved dry land crop production in Vertisols

in Queensland, Australia. In: In Management of Vertisols for

improved agricultural production. Proceeding of an IBSRAM

Inaugural Workshop, February 18–22, 1985, ICRISAT, India.

Cussans, G.W., Raudonius, S., Brain, P., Cumberworth, S., 1996.

Effcets of depth of seed burial and soil aggregate size on seedling

emergence of Alopecurus myosuroides, Galium operine, Stel-

laria media and wheat (Triticum aestivum). Weed Res. 36, 133–

142.

Durr, C., Boiffin, J., 1995. Sugarbeet seedling growth from germi-

nation to first leaf stage. J. Agric. Sci. (Cambridge) 124, 427–

435.

Durr, C., Aubertort, J.N., Richard, G., Dubrulle, P., Duval, Y.,

Boiffin, J., 2001. SIMPLE: A model for simulation of plant

emergence predicting the effect of soil tillage and sowing

operation. Soil Sci. Soc. Am. J. 65, 414–423.

Fidanza, M., Dernoeden, P.H., Zhange, M., 1996. Degree-days for

predicting smooth crab grass emergence in cool season turf.

Crop Sci. 36, 990–996.

Finch-Savage, W.E., Phelps, K., 1993. Onion (Allium cepa L.)

seedling emergence patterns can be explained by the influence

of soil temperature and water potential on seed germination. J.

Exp. Bot. 44, 407–414.

Forcella, F., 1998. Real-time assessment of seed dormancy and seed-

ling growth for weed management. Seed Sci. Res. 8, 201–209.

Forcella, F., Roberto, L., Benech Arnols, Sanchez, R., Ghersa, C.M.,

2000. Modeling seedling emergence. Field Crop Res. 67, 123–

139.


France, J., Thornley, J.H.M., 1984. Mathematical models in agri-

culture and related sciences.. Butterworth, London.

Freebairn, D.M., Wockner, G.H., Silburn, D.M., 1986a. Effect of

catchment management on runoff, water quality and yield

potential from Vertisols. Agric. Water Manage. 12, 1–19.

Freebairn, D.M., Ward, L.D., Clarke, A.L., Smith, G.D., 1986b.

Research and development of reduced tillage systems for Verti-

sols in Queensland, Australia. Soil Till. Res. 8, 211–229.

Gan, Y., Stobe, E.H., Moes, J., 1992. Relative date of wheat seedling

emergence and its impact on grain yield. Crop Sci. 32, 1275–

1281.

Guerif, M., Duke, C.L., Durr, C., 1998. Spatial calibration of a crop

model using optical remote sensing data. A case study on sugar

beet emergence and early growth. In: Proceedings of the First

International Conference on The Geospatial Information in

Agriculture and Forestry Decision Support, Technology and

Applications, Lake Buena Vista, FL, June 1–3, 1998, Erim.

Int., Ann Arbor, MI, pp. 191–197.

Jensen, R.D., 1971. Effect of soil water tension on the emergence

and growth of cotton seedling. Agron. J. 63, 766–768.

Leslie, J.K., 1965. Factors responsible for failure in the establish-

ment of summer grasses of the black earths of the Darling

Downs, Queensland. J. Agril. Anim. Sci. 22, 17–38.

King, C.A., Oliver, L.R., 1994. A model for predicting large

crabgrass (Digitaria sanguinalis) emergence as influenced by

temperature and water potential. Weed Sci. 42, 561–567.

Kremer, E., Lotz, L.A., 1998. Germination and emergence char-

acteristics of triazine-susceptible and triazine-resistant biotypes

of Solanum nigrum. J. Appl. Ecol. 35, 302–310.

Mohanty, M., Painuli, D.K., 2004. Modeling rice seedling emer-

gence and growth under tillage and residue management in a

rice–wheat system on a Vertisol in Central India. Soil Till. Res.

76, 167–174.

Mullins, C.E., Townend, J., Mtakwa, P.W., Payne, C.A., Cowan, G.,

Simmonds, L.P., Daamen, C.C., Dunbabin, T., Naylor, R.E.M.,

1996. Emerge user guide. In: A model to predict crop emergence

in the semiarid topicsunknown:book, Department of Plant and

Soil Science, University of Aberdeen, UK.

Oussible, M., Crookstan, R.K., Larson, W.E., 1992. Subsurface

compaction reduced root and shoot growth and grain yield of

wheat. Agron. J. 84, 34–38.

Radford, B.J., Nielsen, R.G.H., 1983. Extension of crop sowing

times during dry weather by means of stubble mulching and

water injection. Aust. J. Agric. Anim. Husb. 23, 302–308.

Sallaway, M.M., Yule, D.F., Mayer, D., Burger, P.W., 1990. Effect of

surface management on the hydrology of a Vertisol in semi-arid

Australia. Soil Till. Res. 15, 227–245.

Schimpf, D.J., Flint, S.D., Palmblad, I.G., 1977. Representation of

germination curves with the logistic function. Ann. Bot. 41,

1357–1360.

Schneider, E.C., Gupta, S.C., 1985. Predicting corn emergence from

soil temperature, matric potential and aggregate size distribu-

tion. Soil Sci. Soc. Am. J. 49, 415–422.

Sharma, P.K., Verma, T.S., Bhagat, R.M., 1995. Soil structural

improvement with addition of lantana camara biomass in

rice–wheat cropping. Soil Use Manage. 11, 199–203.

Smith, J., Smith, P., Addiscott, T., 1996. Quantitative methods to

evaluate and compare soil organic matter models.. In: Powlson,

D.S., et al. (Eds.), Evaluation of soil organic matter models.

Springer-Verlag, Berlin, pp. 181–199.

Swan, J.B., Kaspar, T.C., Erbach, D.C., Tisdall, T.M., 1996. Seed

row residue management for corn establishment in the Northern

U.S.. Corn Belt. Soil Till. Res. 40, 55–72.

Thomas, G.A., Standley, J., Webb, A.A., Blight, G.W., Hunter,

H.M., 1990. Tillage and crop residue management affect Verti-

sols properties and grain sorghum growth over seven years in

semiarid subtropics. 1. Crop residue and soul water during

fallow periods. Soil Till. Res. 17, 179–181.

Unger, W.P., Ordief, R., 1989. Effective tillage practices for con-

serving soil and water resources. In: Management of Vertisols

for Improved Agricultural Production, Proceeding of an IBS-

RAM Inaugural Workshop, February 18–22, 1985, ICRISAT,

India.

Vleeshouwers L., 1997. Modelling weed emergence patterns. Ph.D.

Dissertation, Wageningen Agricultural University, The Nether-

lands, p. 165.

Estimating impact of puddling, tillage and residue management on wheat seedling emergence and growth...

Documents

Transcript of Estimating impact of puddling, tillage and residue management on wheat seedling emergence and growth...